Bruno Putzeys said:
Those 10000uF BG caps are probably quite a bit more expensive than the modules themselves. The most expensive caps that I have available at home are the 63V 10000uF ELNA Cerafines that are pretty big (5x8cm) and costed something like $30 each. Was planning to use those close to the UcD400 modules as the final power rail decoupling (those ELNAs would be driven by an SMPS with more RCR type of decoupling with cheaper caps, so SMPS + number of RCR parallel caps (0.1Ohm 8200uF 0.1Ohm) and then the ELNAs close to the module.
Are the ELNA`s sonically similar as BG? I`m happy with those ELNA`s in my UcD180 setup so they should be quite OK.
Then the 470uF caps on the module itself maybe interesting as well as these are even closer to the output transistors and may therefore even have a bigger sonic effect than the power supply caps. Was thinking of replacing those with 1000uF 100V 105C types (sticking out since they are higher) or 470uF 100V 105C types. I have 100uF 100V BG caps that fit in there, but going from 470uF to 100uF for those caps may not be such a good idea. Any comments/suggestions?
When I start experimenting to replace those 470uF caps on the UcD module I plan to do some measurements to also have some "objective" info on how they work. I plan to hook-up a scope to measure the voltage over the caps (ringing and drop during switching) and the output voltage of the power stage before the LC filter (also look for ringing and drop during switching, the squarer the pulse the better). These measurements are tricky as normal scope probes are too big and will pick-up other switching noise, so thinking about a solution for this. Now thinking to solder short wires directly to the PCB, twist them and then connect them to a BNC connector and via a BNC cable to the scope. This should pretty much prevent any parasitic EMI/RFI to disturb my measurements. Any tips/ideas/tricks?
Best regards
Gertjan
ghemink said:
Hello Gertjan,
I'd like to know the answer too. That said, I'm quite happy with how my Cerafines sound with my UcD modules.
Cheers, mac.
An externally hosted image should be here but it was not working when we last tested it.
No idea how large Elnas sound. Never tried them. The smaller ones are nice, if a bit euphonic. If you have a couple of large ones to spare uhh well you know the address...ghemink said:
Going to 100uF BG is possible but I can't vouch for their longevity in sustained high-power use. The value of the caps in this location is dictated by their ripple handling capacity, not by their capacitance. Ripple current = approx 0.2 times RMS output current. Not that this helps much. Jelmax is notorious for not specifying anything except a lot of hand-waving.
Bruno Putzeys said:
Hi Bruno,
Thanks for the comments. I have only 8 of those ELNA's and I have to share them between 6 amps (2 woofers amps, 2 mid amps and 2 tweeter amps). So unfortunately for you, I have no spares
. Also not planning to buy more of them as they are quite pricy.One question, that resistor between the 470uF caps and the 10uF caps, is that a 0.005 Ohm resistor? (not 0.05Ohm?). Do you now the value of the inductance L1-L6?
I was not really planning to put a BG instead of the 470uF caps, probably a 1000uF 100V 105C cap will have a lower series resistance and good ripple current. I also have 100V 470uF caps 105C (used before in the SMPS) that also have quite low series resistance. I guess in general 105V caps have the ability to withstand larger ripple currents, is that true?
Best regards
Gertjan
Nice to see Bruno Putzeys has seen the light regarding the big BGs:his description of these caps is identical to Martin Colloms HIFI NEWS test report. I advise any BG skeptic to read his findings in his article "The Blackgate Story" on the AUDIONOTE website,kits & components section.
All my equipment is 90 to 100% Blackgated apart from my recently acquired UCD400 modules, which are operating for the moment as standard , although with the input caps bypassed and fed via T networks with the big BGs in parallel on the outputs of these .... £500 worth even with 19% discount...ouch!
Still they sound pretty damn good and should improve for certain when BG`d further.
ESR & ripple current are the least of your worries when choosing elcaps,the electrolyte is the dominating factor in performance and is where most of the signal losses,noise & distortion occur and that is the secret of BGs:elimination of ion transfer anomilies.
Bob Lewis
All my equipment is 90 to 100% Blackgated apart from my recently acquired UCD400 modules, which are operating for the moment as standard , although with the input caps bypassed and fed via T networks with the big BGs in parallel on the outputs of these .... £500 worth even with 19% discount...ouch!
Still they sound pretty damn good and should improve for certain when BG`d further.
ESR & ripple current are the least of your worries when choosing elcaps,the electrolyte is the dominating factor in performance and is where most of the signal losses,noise & distortion occur and that is the secret of BGs:elimination of ion transfer anomilies.
Bob Lewis
Hmmm...... aren't they discontinued anyway? I hear they are, aren't, are, aren't... ?
"We're not making more, buy them all up while you still can..."
Bruno had they been burnt in at all (takes awhile I hear) or was the sound immediatly an improvement?
Maybe I'll start saving for a pair.
Thanks,
Chris
"We're not making more, buy them all up while you still can..."
Bruno had they been burnt in at all (takes awhile I hear) or was the sound immediatly an improvement?
Maybe I'll start saving for a pair.
Thanks,
Chris
Well we all had to try to find alternatives to 'the one and only', because of the BG end-of-life situation. Listening around many people soon discovered that the alternatives were not so far below the BG's in performance, some even better......
It could be a marketing trick, but not a good one ...!!
All too often. Even "audiophile" brands like marantz pepper their signal paths with coupling electrolytics, only because they are unable or unwilling to invest in DC correct circuitry or trimming.peranders said:
Large analogue mixing consoles are strung together with coupling electrolytics because absense of switching pops is considered more important than sound and the amount of circuitry makes other measures prohibitively expensive. This is one reason why many pop recordings (and quite a few classic and jazz ones) are simply unlistenable.
Annoyingly, studio techs have gotten used to a vague, fuzzy sound and as a result hate mixing on digital desks because suddenly "the mix doesn't come together anymore" (=no intermodulation artefacts to paper over the gaps in the stereo image).
Apart from coupling, also decoupling and power supply capacitors have an embarrassingly noticeable effect on the sound of a circuit. Even with 120dB of PSRR (I've got discrete op amps that do this), the effect is sufficiently large to warrant trying out different alternatives in a listening test.
Bruno,
I know you told here that the Hypex UcD's can be improved a lot compared to how they come out of the box, DIY, ...
I took the jump a while ago, my electronics skills are very basic, and I was able to build myself a power supply. The UcD's themselves, I wouldn't touch with a bargepole, I'm sure if I'd try to change a cap here and there, it would be broken before I knew it.
So, as a "half" DIYer, I'm making sure i'm not buying mass market stuff, but don't go down the super-precision tweaking path. I'd say this forum, and working with UcD's, makes me an informed buyer with an amount of "control" on the quality I want.
With the quote above, does it mean that one could get rid of the input caps, and use another process (without caps in the signal path) to make sure the input is DC-free?
This would mean that, at an added cost, a better UcD module could be possible. I'd be willing to pay for that - I'm also willing to bet I'm not the only one. There's more guys like me, willing to DIY a little to get quality for the money, but too low on skills to fool around with finer electronics.
I know you told here that the Hypex UcD's can be improved a lot compared to how they come out of the box, DIY, ...
I took the jump a while ago, my electronics skills are very basic, and I was able to build myself a power supply. The UcD's themselves, I wouldn't touch with a bargepole, I'm sure if I'd try to change a cap here and there, it would be broken before I knew it.
So, as a "half" DIYer, I'm making sure i'm not buying mass market stuff, but don't go down the super-precision tweaking path. I'd say this forum, and working with UcD's, makes me an informed buyer with an amount of "control" on the quality I want.
With the quote above, does it mean that one could get rid of the input caps, and use another process (without caps in the signal path) to make sure the input is DC-free?
This would mean that, at an added cost, a better UcD module could be possible. I'd be willing to pay for that - I'm also willing to bet I'm not the only one. There's more guys like me, willing to DIY a little to get quality for the money, but too low on skills to fool around with finer electronics.
coupling caps?
With the UcD 400AD modules it is possible to use high quality off board caps by removing and jumpering out the electrolytics. Add the new cap between the input jack and the normal input connector, 2 caps for balanced. This is the same place in the circuit the originals were. Also add a 100k or so resistor across the input jack so any stored charge is bled off. For balanced connectors, resistors go between pin 2 and pin1, pin 3 and pin 1. This should be done anyway even if the original caps are used even though the electrolytics will bleed down by themselves with time.
I can’t speak of this applying to the regular 400 module as I have not seen one yet.
The use of an input cap of high quality like Auricaps is very benign but any DC getting through is defiantly not. For me it is not worth the risk of wiping out the amp and/or speaker. I would recommend 2.2 UF as the value. This gives a long enough time constant to be like DC coupling for even the lowest frequencies. Still short enough to not destroy anything if the preamp should fail. An all solid state solution for direct coupling would have to include DC offset correction and protection to be worry free. For me this is just too complex to consider as the sonic benefits are very questionable.
Roger
With the UcD 400AD modules it is possible to use high quality off board caps by removing and jumpering out the electrolytics. Add the new cap between the input jack and the normal input connector, 2 caps for balanced. This is the same place in the circuit the originals were. Also add a 100k or so resistor across the input jack so any stored charge is bled off. For balanced connectors, resistors go between pin 2 and pin1, pin 3 and pin 1. This should be done anyway even if the original caps are used even though the electrolytics will bleed down by themselves with time.
I can’t speak of this applying to the regular 400 module as I have not seen one yet.
The use of an input cap of high quality like Auricaps is very benign but any DC getting through is defiantly not. For me it is not worth the risk of wiping out the amp and/or speaker. I would recommend 2.2 UF as the value. This gives a long enough time constant to be like DC coupling for even the lowest frequencies. Still short enough to not destroy anything if the preamp should fail. An all solid state solution for direct coupling would have to include DC offset correction and protection to be worry free. For me this is just too complex to consider as the sonic benefits are very questionable.
Roger
Depending on the active circuitry, the power supply decoupling caps are most often also in the signal path, or in the ground return path. Which is about the same as putting them in the actual signal path.
So in that light, most amplifiers use electrolytics in the signal path. And CD players too.
It doesn't matter if you place Marantz in the 'audiophile' class, or (like me) the 'common commercial' class. Most makers in both classes use electrolytics in the signal paths anyway. And sound CAN be improved, simply by changing to better caps.
I do not recommend speaker x-over network caps for this, they are simply not made for this kind of job, and will perform terribly compared to the better (and much cheaper) solutions.
Even BG-NX HiQ is not very good for signal path use.
Using a WIMA MKS2 cap of 2.2 or 4.7 uF is a much better solution. Before you jump to other similar brands, check if the leads are iron or copper. Those with copper leads sound much better than those with iron leads (most brands).
Even better yet is the Vishay SAL 123 RPM series. They just sound GREAT as a DC blocking cap or analog decoupling cap.
So in that light, most amplifiers use electrolytics in the signal path. And CD players too.
It doesn't matter if you place Marantz in the 'audiophile' class, or (like me) the 'common commercial' class. Most makers in both classes use electrolytics in the signal paths anyway. And sound CAN be improved, simply by changing to better caps.
I do not recommend speaker x-over network caps for this, they are simply not made for this kind of job, and will perform terribly compared to the better (and much cheaper) solutions.
Even BG-NX HiQ is not very good for signal path use.
Using a WIMA MKS2 cap of 2.2 or 4.7 uF is a much better solution. Before you jump to other similar brands, check if the leads are iron or copper. Those with copper leads sound much better than those with iron leads (most brands).
Even better yet is the Vishay SAL 123 RPM series. They just sound GREAT as a DC blocking cap or analog decoupling cap.
Attachments
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.